Corpus ID: 219687524

COT-GAN: Generating Sequential Data via Causal Optimal Transport

@article{Xu2020COTGANGS,
  title={COT-GAN: Generating Sequential Data via Causal Optimal Transport},
  author={Tianlin Xu and L. Wenliang and M. Munn and B. Acciaio},
  journal={ArXiv},
  year={2020},
  volume={abs/2006.08571}
}
We introduce COT-GAN, an adversarial algorithm to train implicit generative models optimized for producing sequential data. The loss function of this algorithm is formulated using ideas from Causal Optimal Transport (COT), which combines classic optimal transport methods with an additional temporal causality constraint. Remarkably, we find that this causality condition provides a natural framework to parameterize the cost function that is learned by the discriminator as a robust (worst-case… Expand
Conditional Versus Adversarial Euler-based Generators For Time Series
A Data-Driven Market Simulator for Small Data Environments
Generative Adversarial Network: Some Analytical Perspectives
Quantized Conditional COT-GAN for Video Prediction
Conditional and Adversarial Euler-based Generators For Time Series
Cournot-Nash Equilibrium and Optimal Transport in a Dynamic Setting
The Universal Approximation Property

References

SHOWING 1-10 OF 51 REFERENCES
Time-series Generative Adversarial Networks
Learning Generative Models with Sinkhorn Divergences
Improving GANs Using Optimal Transport
Real-valued (Medical) Time Series Generation with Recurrent Conditional GANs
Demystifying MMD GANs
Generative Adversarial Nets
f-GAN: Training Generative Neural Samplers using Variational Divergence Minimization
Disentangled Sequential Autoencoder
...
1
2
3
4
5
...